Overprescribing antidepressants to children: pharmacoepidemiological study in primary careBMJ 2005; 331 doi: https://doi.org/10.1136/bmj.331.7530.1451 (Published 15 December 2005) Cite this as: BMJ 2005;331:1451
- Kathleen Bennett, statistician ()1,
- Mary Teeling, lecturer in clinical pharmacology1,
- John Feely, professor of pharmacology and therapeutics1
- 1 Department of Pharmacology and Therapeutics, Trinity Centre for Health Sciences, St James's Hospital, Dublin 8, Republic of Ireland
- Correspondence to: K Bennett
- Accepted 16 August 2005
Depression is associated with considerable morbidity and risk of suicide. About 20% of adults and 2% of those aged 0-18 years are affected at anytime.1 Although widely used to manage adult depression, no antidepressants are licensed for use in children. Recent attention has focused on the potential risk of suicide in children. We aimed to evaluate the level of use of antidepressants in adults and children in Ireland, the drug type and duration of use, and the changes in prescription rates over time.
Participants, methods, and results
We used the General Medical Services' claims database. The scheme, which is means tested, provides free health services to about 30% of the Irish population (representing 1.24 million people) and 28% of all children.2 All prescription items are coded using WHO's anatomical therapeutic chemical classification (ATC) and basic demographic information (age and sex) are recorded. As age is categorised we classified people aged 0-15 years and ≥ 16 years as children and adults respectively. We identified people who had been prescribed antidepressants (ATC N06A-X) between January 2001 and August 2004. Prevalence was based on 2003 data. We used Poisson regression to examine trends, expressed as the average monthly relative change in prescription rate (that is, a rate ratio of 0.98 represents as a relative rate change of −2%). We used SAS 9.0 (SAS, Cary, North Carolina) for all analyses.
In 2003, antidepressants were prescribed to 1079 children, representing 0.43% of the eligible population, and to 153 863 adults, representing 16.9% of the population (ages 16-24, 9%; 25-74, 17.9%; ≥ 75 years, 18.3%). Girls (odds ratio 1.60, 95% confidence interval 1.42 to 1.80) and women (1.40, 1.38 to 1.42) were significantly more likely to receive antidepressants than their male counterparts. Type of antidepressant was similar in adults and children, with selective serotonin reuptake inhibitors the most commonly prescribed group. About 58% of children received only one prescription for antidepressants, but 19.4% received three or more months of prescriptions in 2003. In contrast, 23.5% of adults received just one prescription, and 66.5% received three or more months' worth, reflecting more chronic use.
The test for interaction of time by population group (adult or child) was significant (P < 0.001), indicating different trends for the two groups, as was the interaction of time by type of antidepressant (selective serotonin reuptake inhibitor versus tricyclics) for children (P < 0.001).
The overall trend in antidepressant prescribing in children showed a significant reduction between January 2001 and August 2004 (figure; rate ratio −0.45%, 95% confidence interval −0.60% to −0.30%, P < 0.001). A similar trend was noted for tricyclic antidepressants (−1.80%, −2.10% to −1.50%). However prescribing patterns for selective serotonin reuptake inhibitors in children did not show an overall downward trend (0.17%, −0.03% to 0.37%, P = 0.091). In contrast, the adult pattern showed a statistically significant upward trend (1.06%, 1.05% to 1.07%, P < 0.001).
What is already known on this topic
Although antidepressants are widely used to manage depression in adults, no agents are licensed for use in children, and risk of suicide is a concern
What this study adds
Prescribing selective serotonin reuptake inhibitors in adults to children has not significantly reduced despite safety warnings
We found a significant decrease in antidepressant prescribing in children over time (except for selective serotonin reuptake inhibitors), whereas prescribing increased in adults. Most children received only one month's treatment, although such use may have limited efficacy.3
Several limitations to our study may have biased our findings. Diagnostic data were not available and social disadvantage (a factor associated with depression and other morbidities) is over-represented in the database. Such children will often have reduced access to specialist health care, potentially necessitating pharmacotherapy in the short term. The results suggest that selective serotonin reuptake inhibitors continue to be prescribed for the management of childhood depression, despite warnings from regulatory authorities.4 5 Prescribers may need to receive regular reminders about potential safety problems and efficacy with use of antidepressants in children, especially in the first weeks of treatment, to ensure their appropriate use.
We thank the General Medical Services (Payments) Board for the use of the prescribing database.
Papers p 1435
Contributors KB and MT planned, designed, analysed, and drafted the article. KB, MT, and JF interpreted and approved the final version. KB is guarantor.
Ethical approval Not needed.
Competing interests None declared.